RHELM (Long Context) - Mixed: leaderboard
RHELM category (Zhang et al., Microsoft and Renmin University of China, 2026) that needs dialogue history and documents together, such as finding content next to a discussed topic or analysing a document after the edits discussed in conversation. The model reads the whole history, dialogue and external documents in chronological order, in a 1M-token context; accuracy in percent on 210 questions judged 0 or 1 by an LLM, higher is better.
Metric: Accuracy (%; 210 mixed dialogue-and-document questions, LLM judge, dialogue history plus external sources, full history in a 1M-token context). Source: arxiv.org. Saturation forecast: Around July 2028. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash Lite | 13.1 | #958 |
| 2 | Qwen2.5-14B-Instruct-1M | 6.2 | #1417 |
| 3 | GPT-4.1 Mini | 0.3 | #775 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=rhelm-long-context-mixed · How It Works · Data refreshed daily, snapshot 2026-11-12.